INDEX

Explanations

terms related to technical or professional topics, possibly related to software development or project management

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

asio

-0.91

ESE

-0.89

okia

-0.89

EStream

-0.89

ETS

-0.88

YS

-0.86

SHIP

-0.81

iage

-0.80

PsyNetMessage

-0.77

ICES

-0.77

POSITIVE LOGITS

 flavorful

1.05

 adventurous

1.04

 unpredictable

1.03

 economical

1.02

 efficient

1.00

 unpre

1.00

 witty

0.98

 pragmatic

0.98

 imaginative

0.97

 uncomp

0.96

Activations Density 1.765%

terms related to technical or professional topics, possibly related to software development or project management

No Comments

No Known Activations